Optimization
The fifth component necessary for building a deep learning model. The optimization algorithm to take minibatches of data, run them through the model, and minimize loss.
At a minimum, the optimization algorithm will take a dataset, a loss, and a model. You can then run the .train() method to update the model's parameters to minimize loss over the dataset. Please refer to the package documentation to find out about the different optimization methods.
Updated less than a minute ago